By Lina Irawan 
On February 20, the technology landscape experienced a seismic tremor that revealed a profound and growing chasm between the intricate realities of specialized innovation and the often-simplistic valuation mechanisms of public markets. The catalyst was Anthropic’s unveiling of Claude Code Security, an AI-powered code-scanning tool designed to enhance developer workflows and identify vulnerabilities earlier in the software development lifecycle. By the close of the trading day, a staggering sum—billions of dollars—had evaporated from the market capitalization of prominent cybersecurity stocks, sending shockwaves through the sector. Industry giants bore the brunt of this irrational exuberance: CrowdStrike, a leader in endpoint protection and cloud security, saw its stock tumble by 10%. Zscaler, a pioneer in zero-trust network access and cloud security, experienced an even sharper decline of 11%. Okta, a dominant force in identity and access management, shed 9% of its value.
Yet, a closer examination of Anthropic’s announcement reveals a stark disconnect between the actual product launch and the market’s visceral reaction. Anthropic did not introduce an endpoint protection platform (EPP), which safeguards individual devices like laptops and servers from threats. It did not unveil an identity access and management (IAM) platform, crucial for verifying user identities and controlling access to digital resources. Crucially, it did not present an alternative to the foundational zero-trust architecture, a strategic cybersecurity model that assumes no user or device should be trusted by default, regardless of whether they are inside or outside the organization’s perimeter. Instead, Anthropic released a new capability within Claude Code, its existing AI-powered developer tool—a highly specialized application security offering focused on code analysis. The market, however, reacted as if an entire, diverse sector had been fundamentally and structurally disrupted by a single, albeit innovative, feature release.
My initial reaction to the selloff wasn’t one of surprise at investor irrationality; market overreactions are a common phenomenon. What truly struck me was the underlying assumption necessary for such a dramatic shift: that the labels “AI” and “cybersecurity” are interchangeable. This conflation represents a critical misunderstanding of the modern technology landscape. Endpoint protection, identity access and management, network security, application security, and developer tooling are not merely different facets of a single, monolithic cybersecurity domain. They are distinct, deeply specialized disciplines, each characterized by unique architectural requirements, targeted buyers (ranging from security operations teams to development teams), and fundamentally different economic models. Professionals building within this space grasp these distinctions instinctively. The market’s failure to do so, however, tells a far more interesting and concerning story about the limitations of our current financial infrastructure.
For decades, public markets have been structured around the paradigm of the generalist investor. Portfolio managers and analysts have been expected to cover an enormous intellectual territory, shifting from cloud infrastructure one day, to fintech the next, and semiconductors the day after. This model functioned adequately when industries were broader, less specialized, and evolved at a more glacial pace. However, technology no longer adheres to this antiquated rhythm. What we collectively label “tech” today is not a singular, homogeneous sector. It is a vibrant, intricate collection of deeply specialized domains, each operating with its own unique economic drivers, competitive dynamics, and technological underpinnings. The expertise required to accurately value a cutting-edge cybersecurity company, with its complex threat landscape and compliance requirements, is fundamentally different from the expertise needed to assess an AI infrastructure company, with its demands for massive computational power and proprietary algorithms. Yet, in the public markets, the portfolio manager responsible for allocating capital across both these disparate domains is often the same individual.
This situation finds a compelling analogy in other established markets. We would never conceive of asking a commodities trader to price crude oil, copper, and wheat as if they were merely variations of the same underlying asset. These markets long ago evolved specialized exchanges, dedicated analysts, and sophisticated pricing structures tailored to the unique characteristics of each commodity. In the technology sector, however, we persist in the pretense that the generalist model remains sufficient, despite the overwhelming evidence to the contrary.
The AI Narrative: A Double-Edged Sword of Confusion and Hype
The prevailing AI narrative, with its pervasive hype and often exaggerated immediate impact, exacerbates this problem considerably. Wall Street appears to be pricing AI as if it has already fundamentally reshaped the global economy, driving unprecedented productivity gains and widespread disruption. However, real-world data paints a much more nuanced picture. A comprehensive survey published by the National Bureau of Economic Research (NBER) in February, gathering insights from nearly 6,000 executives across the U.S., UK, Germany, and Australia, revealed a striking reality: over 80% of respondents reported zero measurable impact of AI on their company’s productivity or employment over the preceding three years. While AI’s long-term potential for transformation is undeniable and will undoubtedly be profound, the current gap between what the market is pricing in—often reflecting speculative enthusiasm—and what is genuinely happening inside companies in terms of tangible, widespread operational impact, remains enormous. This chasm between perception and reality makes rational valuation even more challenging for generalist investors attempting to navigate a specialized landscape through a broad AI lens.
Compounding these issues, the center of gravity in capital markets has undergone a dramatic shift over the last two decades, moving decisively towards private ownership. Larry Fink, CEO of BlackRock, highlighted this trend in his influential 2025 annual letter to investors, noting that a staggering 81% of U.S. companies with annual revenues exceeding $100 million are now privately held. This dramatic shift is mirrored by the precipitous decline in the number of publicly traded companies, which has fallen by approximately 50% since the 1990s. While common explanations often point to the burdens of regulatory compliance or the relentless pressure of quarterly earnings reporting, these factors, while relevant, are arguably secondary. The deeper, more systemic issue is that public markets, in their current generalist structure, lack the sophisticated machinery and specialized expertise required to properly value complex, rapidly evolving technology companies.
This phenomenon is strikingly evident at the apex of the tech industry. Consider OpenAI, which recently secured a staggering $110 billion in funding at a $730 billion valuation, all while remaining a private entity. Stripe, a financial technology powerhouse, continues to prioritize long-term growth and strategic positioning over an immediate public offering. Databricks, a leader in data and AI, is scaling to multi-billion-dollar revenues while steadfastly maintaining its private status. These companies are not actively avoiding scrutiny; on the contrary, they are under intense scrutiny from highly sophisticated private investors. What they are strategically avoiding is the inherent risk of mispricing and misunderstanding that often accompanies a public listing in a generalist market.
This trend has fostered the emergence of a vast, influential tier of companies that operate above the traditional venture capital stage but remain below the public markets. These are enterprises generating substantial revenue, operating at significant scale, and exerting global impact. This dynamic layer barely existed two decades ago, representing a structural change in how and where value is created and captured in the technology ecosystem.
Rethinking Technology Exposure: A Path Towards Specialization
Given these profound shifts, it is imperative to rethink how we organize and approach technology exposure in capital markets. One proposed idea, and I readily acknowledge its incompleteness, is to move beyond a single, broad “tech” umbrella, analyzed predominantly by generalists, towards a more hierarchical and specialized structure. At the highest level, asset allocators would make strategic decisions about how much capital to deploy into broad categories like cybersecurity, AI infrastructure, fintech, or vertical SaaS. Below this, each distinct domain would foster its own ecosystem of specialist analysts, develop bespoke valuation models tailored to its specific economics and competitive landscape, and potentially even have its own indices built around its unique realities.
Under such a framework, cybersecurity would undeniably stand on its own, not merely as a marketing bucket, but as a robust analytical category staffed by experts who possess a deep, nuanced understanding of how these businesses operate. Similarly, AI infrastructure would warrant its own dedicated analytical domain. Would such a restructuring eliminate market volatility? Of course not; markets will always be prone to chasing the next compelling narrative. However, it would significantly reduce the kind of blind correlation and indiscriminate selloffs we witnessed on February 20, where fundamentally different companies, offering disparate solutions to distinct problems, moved in lockstep simply because they shared a generic headline.
The consequences of the current generalist model extend far beyond mere stock price fluctuations. When founders and management teams perceive that public markets cannot adequately differentiate between their highly specialized business and a seemingly similar but fundamentally different competitor, they adjust their strategies. They choose to remain private for extended periods, opt to raise additional rounds of private capital, and increasingly seek liquidity through alternative channels. The capital is undeniably available—in many cases, more abundant than ever before—but it increasingly resides behind closed doors, within the private equity and venture capital ecosystems.
This fundamental shift has profound implications for the broader economy and for individual investors. If the defining technology companies of our era continue to scale and mature predominantly outside the public markets, most individual investors, pension funds, and mutual funds will only gain access to these companies once the heavy lifting of innovation, market validation, and exponential growth has largely been completed, or, more often, not at all. As a result, the most significant upside—the exponential returns associated with early-stage growth and market creation—concentrates quietly, primarily in the hands of sovereign wealth funds, large institutional investors, and mega-venture capital firms.
This concentration of returns doesn’t merely shift who benefits; it fundamentally alters the economics of investing itself. As a greater proportion of investment returns is captured within private markets, scale and access become structural advantages, reinforcing the dominance of the largest pools of capital. Concurrently, it reduces the opportunity set and potential returns available in public markets. Individual investors, along with the vast network of pension funds and mutual funds that largely remain anchored to public markets, are left competing over a shrinking share of growth and innovation, thereby altering return expectations across the entire financial system.
February 20, therefore, wasn’t truly about Anthropic’s specific product or even solely about the cybersecurity sector. It was a stark symptom of a larger, systemic issue: a headline triggered a widespread reaction across an entire category, despite the companies within it performing vastly different functions. This growing gap between the increasing specialization of technology and the largely unchanged, generalist approach to grouping and pricing it in public markets is becoming impossible to ignore. A fundamental re-evaluation of how technology investments are categorized and assessed in the public sphere is not merely advisable; it is increasingly essential for the health and accessibility of capital markets.
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.